2.0.3 planning

[Tollef Fog Heen]

Hi,

I'm evaluating what changes that should go into 2.0.3 and have come up
with the following list:

------------------------------------------------------------------------
r3433 | phk | 2008-11-25 09:37:34 +0100 (ti., 25 nov. 2008) | 6 lines

When we receive an If-Modified-Since on an ESI object, do not process the conditional
for the child object and pretend to send a 304 reply for them, if we have decided to
deliver the main object.

Fixes #386

------------------------------------------------------------------------
r3361 | tfheen | 2008-11-06 12:46:31 +0100 (to., 06 nov. 2008) | 7 lines

Fix up $N vs \N in man page

The VCL man page documented the capturing parentheses as using $N
rather than \N which is actually used.

Fixes #359

------------------------------------------------------------------------
r3362 | tfheen | 2008-11-06 12:57:05 +0100 (to., 06 nov. 2008) | 4 lines

Document the size parameter to -s malloc

Fixes #362

------------------------------------------------------------------------
r3366 | phk | 2008-11-10 10:29:52 +0100 (ma., 10 nov. 2008) | 12 lines

Add a debug CLI command to seed random(3).

This is a lot less useful than it could have been, as the Open Group
only mandates that:

        Like rand(), random() shall produce by default a sequence
        of numbers that can be duplicated by calling srandom() with
        1 as the seed.

But crucially leaves out *which* sequence of numbers.

------------------------------------------------------------------------
r3367 | phk | 2008-11-10 10:37:21 +0100 (ma., 10 nov. 2008) | 14 lines

Add a toplevel word which examines the sequence returned by
srandom(1) and stops the test if we do not get the same sequence
as we expect.

The Open Group does not define which deterministic sequence srandom(1)
should result in, on that it be deterministic, but I have high hopes
in the general sanity and expect that UNIX people across the board
have realized that for portability the same sequence should be
returned on all platforms.

At the very least FreeBSD and Linux/GLIBC, as seen on projects.linpro.no,
agree.

------------------------------------------------------------------------
r3368 | phk | 2008-11-10 10:40:39 +0100 (ma., 10 nov. 2008) | 7 lines

Add a test of the random director that uses actual randomness.

This will be skipped on platforms where srandom(1) does not result
in the same deterministic sequence of random numbers as on FreeBSD
and Linux.

------------------------------------------------------------------------
r3386 | phk | 2008-11-11 20:06:55 +0100 (ti., 11 nov. 2008) | 5 lines

Implement restart in vcl_hit.

Fixes #365

------------------------------------------------------------------------
r3387 | phk | 2008-11-11 20:07:30 +0100 (ti., 11 nov. 2008) | 3 lines

Regression test case for ticket 365: restart in hit.


------------------------------------------------------------------------
r3388 | phk | 2008-11-11 21:22:05 +0100 (ti., 11 nov. 2008) | 8 lines

Make sure that
        set obj.ttl = 0s
means that the object is not hit again by actually using "-1" instead.

This works around the rounding error which otherwise causes the object
to be inside TTL for up to one second - epsilon.

------------------------------------------------------------------------
r3401 | tfheen | 2008-11-18 14:29:34 +0100 (ti., 18 nov. 2008) | 5 lines

Make malloc print max storage size

storage_file prints the maximum storage size, make malloc do the same,
for consistency.

------------------------------------------------------------------------
r3402 | tfheen | 2008-11-18 21:53:03 +0100 (ti., 18 nov. 2008) | 5 lines

Document grace

Thanks to perbu for suggested documentation
Fixes 355

------------------------------------------------------------------------
r3407 | tfheen | 2008-11-19 17:26:16 +0100 (on., 19 nov. 2008) | 4 lines

Correct defaults in varnishd.1

The defaults for thread_pool_min and thread_pools were wrong; fixed.

------------------------------------------------------------------------
r3408 | phk | 2008-11-20 09:50:56 +0100 (to., 20 nov. 2008) | 3 lines

Check ECONNRESET

r3418 | tfheen | 2008-11-22 02:35:16 +0100 (lø., 22 nov. 2008) | 2 lines

Fix typo

------------------------------------------------------------------------
r3420 | phk | 2008-11-24 11:05:55 +0100 (ma., 24 nov. 2008) | 5 lines

Update license to remove the advertising clause, reflecting similar change
in the FreeBSD original.

Approved by:    des
------------------------------------------------------------------------
r3426 | tfheen | 2008-11-24 15:04:42 +0100 (ma., 24 nov. 2008) | 4 lines

Fix typo (s/timeout/interval/) in default parameters for backend health

Thanks to Jonny @ globo for noticing.

------------------------------------------------------------------------
r3463 | ingvar | 2008-12-18 10:47:05 +0100 (to., 18 des. 2008) | 1 line

Changed rpm summary string as requested by the Fedora project
------------------------------------------------------------------------
r3466 | phk | 2008-12-18 11:31:08 +0100 (to., 18 des. 2008) | 5 lines

Add a missing error check.

Fixes #409

------------------------------------------------------------------------
r3488 | phk | 2008-12-21 19:33:30 +0100 (sø., 21 des. 2008) | 3 lines

Assert fcntl(2) have not failed.

------------------------------------------------------------------------
r3490 | phk | 2008-12-21 19:34:02 +0100 (sø., 21 des. 2008) | 3 lines

Assert that realloc() did.

------------------------------------------------------------------------
r3492 | phk | 2008-12-21 19:35:24 +0100 (sø., 21 des. 2008) | 4 lines

Be more stringent about our timeout: fail even if we did get a
socket, but took to.

------------------------------------------------------------------------

I'm also wondering if we should merge the «backend timeout patch»,
namely 3406, 3411, 3412:

r3406 | petter | 2008-11-19 15:13:57 +0100 (on., 19 nov. 2008) | 8 lines

Added support for setting read timeouts for backend requests
(first_byte_timeout and between_bytes_timeout), in addition to make the
connect_timeout available for the bereq object in vcl_miss and
vcl_fetch.

first_byte_timeout is a read timeout from the connection to the backend
is created to when the first byte arrives. It can be set as a parameter
to varnish, as a field in the backend declaration or as
bereq.first_byte_timeout in vcl_miss and vcl_pass.

between_bytes_timeout is a read timeout between each read from the
backend. It can be set as a parameter to varnish, as a field in the
backend declaration or as bereq.between_bytes_timeout in vcl_miss and
vcl_pass.

The time unit for these timeout values are seconds. NOTE: The
connect_timeout previously used milliseconds as time unit, so beware.

------------------------------------------------------------------------
r3411 | petter | 2008-11-20 12:01:26 +0100 (to., 20 nov. 2008) | 1 line

Added documentaiton on the timeouts
------------------------------------------------------------------------
r3412 | petter | 2008-11-21 08:21:13 +0100 (fr., 21 nov. 2008) | 2 lines

Soooorry.

Comments?  Thoughts?

-- 
Tollef Fog Heen 
Redpill Linpro -- Changing the game!
t: +47 21 54 41 73